This paper analyzes the determinants of kindergarten (KG) private school choice and child learning outcomes in Ghana, the first country in sub-Saharan Africa to expand two years of pre-primary school as part of the universal public education system. Using data from a two-year experimental intervention implemented in the Greater Accra Region in the 2015-16 school year, we examine the role of child, household, school, teacher, and classroom-level characteristics in predicting public vs private preschool choice, and explore whether parental characteristics affect test scores beyond the school selection channel. We find evidence of marked public-private differences in child, household, and school characteristics, largely favoring private schools. Furthermore, we show that parental characteristics are associated with baseline child test scores beyond their influence on school choice. Parental involvement – such as number of books at home and parent’s help on homework – is most predictive of school readiness, suggesting that “investments” in children go beyond simply selecting the type of school presumed to be best. Lastly, we find evidence of an independent effect of private school education on follow-up early literacy scores. Our findings contribute to knowledge on the recent expansion of early childhood education and private schools globally and in sub-Saharan Africa.